Grok-2 Mini: Complete Benchmark Performance Guide

xai 2-mini 2024-08-13 Proprietary

In-depth analysis of Grok-2 Mini performance on MMLU, HumanEval, GSM8K, MATH, BBH benchmarks, including pricing and use cases.

Key Specifications

Vendorxai
Version2-mini
Release Date2024-08-13
Context Window131072 tokens
Input Modalitiestext
Output Modalitiestext
LicenseProprietary
Documentationhttps://docs.x.ai/

Benchmark Performance

BenchmarkScoreUnitEvaluated AtNotesSource
MMLU80.6%2024-08-135-shotview
HUMANEVAL77.9pass@12024-08-13—view
GSM8K83.6%2024-08-130-shot CoTview
MATH45.1%2024-08-130-shot CoTview
BBH74.6%2024-08-133-shot CoTview
GPQA31.7%2024-08-130-shotview
IFEVAL73.5%2024-08-13prompt_strictview
ARC92.2%2024-08-13challengeview
MUSR51.4%2024-08-130-shotview
WINOGRANDE84.7%2024-08-130-shotview

Pricing

TierPriceCurrency
Input$0.1 / MtokUSD
Output$0.4 / MtokUSD
Cache Read$0 / MtokUSD
Cache Write$0 / MtokUSD

Source: https://x.ai/api · as of 2024-08-13

Compliance

  • Data Residency: US
  • SOC2: ✗
  • HIPAA: ✗
  • GDPR: ✗
  • ISO 27001: ✗

Grok-2 Mini

Ikhtisar Model

xAI Grok-2 Mini 经济型模型, 131K 上下文, 速度快成本低, 适合实时对话与高吞吐场景。

Spesifikasi Inti

VendorVersiTanggal RilisJendela KonteksModalitas InputModalitas OutputLisensi
Xai2-mini2024-08-13131KtexttextProprietary

Kinerja Benchmark

BenchmarkSkorSatuanCatatan
MMLU (Massive Multitask Language Understanding)80.6%5-shot
HumanEval77.9pass@1—
GSM8K (Grade School Math 8K)83.6%0-shot CoT
MATH45.1%0-shot CoT
BBH (BIG-Bench Hard)74.6%3-shot CoT
GPQA31.7%0-shot
IFEval73.5%prompt_strict
ARC92.2%challenge
MUSR51.4%0-shot
WinoGrande84.7%0-shot

Harga

InputOutputBaca CacheTulis Cache
————

per juta token

Kelebihan

  • MMLU score 80.6, strong knowledge reasoning.

Kekurangan

  • 闭源专有模型，不支持自托管。

Kasus Penggunaan

  • 代码生成与调试

Referensi